Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    nominal

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     merges
    -0.07
    jang
    -0.07
    -action
    -0.06
     Juan
    -0.06
    eyn
    -0.06
    uran
    -0.06
    -shaped
    -0.06
     dangerously
    -0.06
     невозможно
    -0.06
     luck
    -0.06
    POSITIVE LOGITS
     nominal
    0.08
     اسم
    0.07
    _Ref
    0.07
    kového
    0.06
    mentor
    0.06
    annies
    0.06
    eral
    0.06
     sala
    0.06
     offre
    0.06
    Internal
    0.06
    Act Density 0.002%

    No Known Activations